Search in Elastic Search with AND operator - elasticsearch

Question

I have the following documents in my ES
name type
Thomas Robert King xyz
Peter Jacob Junior xyz
Robert Einstein Jr abc
Thomas Jupyin DR xyz
I want to get all the documents where it matches for Robert and Thomas both in a sentence and are of type xyz.
The current ES query that i have created is :
http://localhost/index/type/_search/?size=100&pretty=1/
{
"query": {
"bool": {
"must": [
{
"match":
{ "name": "Thomas" }
},
{
"match":
{ "name": "Robert" }
}
],
"must_not": [
{
{
"match":
{ "type": "abc" }
}
]
}
}
}
Is it the right way to create a AND scenario in elastic search or is there a better query to handle the same output .
I am using ES 2.1
Thanks

Answer 1

Given your current goal: get all the documents where it matches for Robert and Thomas both in a sentence and are of type xyz, you should not use a must_not query on abc. If there is another type ( ubv for example), your query will also get the documents which match this type.
You can do that query:
{
"query": {
"bool": {
"must": [
{
"match":
{
"name": {
"query": "Thomas Robert",
"operator": "and"
}
}
},
{
"match":
{
"type": "xyz"
}
}
]
}
}
}
By the way, the clauses must are scoring query in a bool query. If you want just want to filter on type xyz you should use:
{
"query": {
"bool": {
"must": [
{
"match":
{
"name": {
"query": "Thomas Robert",
"operator": "and"
}
}
}],
"filter": [
{
"match":
{
"type": "xyz"
}
}
]
}
}
}
Moreover please look at term level query versus full-text query, to see what you really need. If your "type" field is not analyzed, you should use term query instead of match to avoid analysis in match query.
https://www.elastic.co/guide/en/elasticsearch/reference/2.3/term-level-queries.html